Introducere: Previous studies have shown that obe-sity is associated with a lower mortality and lower N-Terminal pro Brain Natriuretic Peptide (NTproBNP) levels, when compared to patients with normal body mass index (BMI) in chronic heart failure. However, there are conflicting data regarding this paradox in acute heart failure (AHF).
Objective: The purpose of the study was to define the relationship between obesity, NTproBNP, and long-term outcomes in AHF with reduced ejection fraction (HFrEF) vs. non-reduced ejection fraction (HFnrEF). Methods: We performed a retrospective analysis of all patients diagnosed with AHF between 2013 and 2015. Patients were classified into 2 groups: HFrEF (LVEF <40%) and HFnrEF (LVEF ≥40%). Each group was subdivided in non-obese (BMI <30 kg/m²) (NO) and obese patients (BMI ≥30 kg/m²) (O). All patients had data on comorbidities, and a standard biological evalu-ation. After adjusting for age, sex and presence of dia-betes, we compared NTproBNP level in each subgroup. The evaluated outcomes were 2-year mortality and 2-year composite outcome (CO) (mortality, hospitali-zation for HF, acute coronary syndrome or stroke).
Results: 241 patients had complete data for analysis (36-92 years), 139 with HFrEF and 102 with HFnrEF. O patients were significantly younger than NO patients in both groups (HFrEF p=0.01, HFnrEF p=0.05). There was no difference in NTproBNP levels in HFrEF group, between O and NO patients (8464±1266 pg/ml vs. 8737±1295 pg/ml, p=0.39). However, in HFnrEF group, NTproBNP was significantly lower in O vs. NO patients (3771±1262 pg/ml vs. 7846±1054 pg/ml, p=0.013). In HFrEF group, the 2-year CO was significantly lower in NO patients (44.3% vs. 64.5%, p=0.014). Moreo-ver the 2-year mortality was significantly lower in O vs. NO patients (42.6% vs. 59.2%, p=0.039). This paradox was not observed in the HFnrEF group. By univaria-te analysis, NTproBNP did not predict mortality and CO in HFrEF obese patients and in all HFnrEF group. However, in NO patients with HFrEF, NTproBNP predicts mortality (R=0.34, R2=0.14, p=0.027) and CO (R=0.40, R2=0.16, p=0.008).
Conclusions: In AHF, the obesity paradox seems to be present only in HFrEF, with lower mortality and mor-bidity for these patients. NTproBNP has a prognostic value only in NO but not in O patients with HFrEF. In obese patients with HFnrEF, NT-proBNP levels are not related to the severity of cardiac dysfunction. New diagnostic cutoff value for NTproBNP may be needed in this subgroup of patients.